inmo-2012 previous year question papers of indian national mathematical olympiad (inmo) with...

7/29/2019 inmo-2012 Previous year Question Papers of Indian National Mathematical Olympiad (INMO) with solutions

1/6

Problems and Solutions: INMO-2012

1. Let ABCD be a quadrilateral inscribed in a circle. Suppose AB =2 +

2 and AB subtends 135 at the centre of the circle. Find the

maximum possible area of ABCD.

Solution: Let O be the centre of the circle in which ABCD is inscribedand let R be its radius. Using cosine rule in triangle AOB, we have

2 +

2 = 2R2(1

cos 135) = R2(2 +

2).

Hence R = 1.

Consider quadrilateral ABCD as in the second figure above. Join AC.For [ADC] to be maximum, it is clear that D should be the mid-pointof the arc AC so that its distance from the segment AC is maximum.Hence AD = DC for [ABCD] to be maximum. Similarly, we concludethat BC = CD. Thus BC = CD = DA which fixes the quadrilateralABCD. Therefore each of the sides BC, CD, DA subtends equal anglesat the centre O.

LetBOC = , COD = and DOA = . Observe that

[ABCD] = [AOB]+[BOC]+[COD]+[DOA] =1

2sin135+

1

2(sin +sin +sin ).

Now [ABCD] has maximum area if and only if = = = (360 135)/3 = 75. Thus

[ABCD] =1

2sin 135 +

3

2sin75 =

1

2

1

2+ 3

3 + 1

2

2

=

5 + 3

3

4

2.

Alternatively, we can use Jensens inequality. Observe that , , are

all less than 180

. Since sin x is concave on (0, ), Jensens inequalitygives

sin + sin + sin

3 sin

+ +

3

= sin 75.

Hence

[ABCD] 12

2+

3

2sin75 =

5 + 3

3

4

2,

with equality if and only if = = = 75.

e JEE Mains/JEE Advanced Online Prep and Guidance on www.100Marks.in


2/6

2. Letp1 < p2 < p3 < p4 and q1 < q2 < q3 < q4 be two sets of prime numberssuch that p4 p1 = 8 and q4 q1 = 8. Suppose p1 > 5 and q1 > 5. Provethat 30 divides p1 q1.Solution: Since p4 p1 = 8, and no prime is even, we observe that{p1, p2, p3, p4} is a subset of{p1, p1 + 2, p1 + 4, p1 + 6, p1 + 8}. Moreoverp1 islarger than 3. If p1 1 (mod 3), then p1 + 2 and p1 + 8 are divisible by 3.Hence we do not get 4 primes in the set

{p1, p1 + 2, p1 + 4, p1 + 6, p1 + 8

}.

Thus p1 2 (mod 3) and p1 + 4 is not a prime. We get p2 = p1 + 2, p3 =p1 + 6, p4 = p1 + 8.

Consider the remainders of p1, p1 + 2, p1 + 6, p1 + 8 when divided by 5. Ifp1 2 (mod 5), then p1 + 8 is divisible by 5 and hence is not a prime. Ifp1 3 (mod 5), then p1 + 2 is divisible by 5. If p1 4 (mod 5), then p1 + 6is divisible by 5. Hence the only possibility is p1 1 (mod 5).Thus we see that p1 1 (mod 2), p1 2 (mod 3) and p1 1 (mod 5). Weconclude thatp1 11 (mod 30).Similarly q1 11 (mod 30). It follows that 30 divides p1 q1.

3. Define a sequence f0(x), f1(x), f2(x), . . . of functions byf0(x) = 1, f1(x) = x,

fn(x)

2 1 = fn+1(x)fn1(x), for n 1.Prove that each fn(x) is a polynomial with integer coefficients.

Solution: Observe that

f2n

(x) fn1(x)fn+1(x) = 1 = f2n1(x) fn2(x)fn(x).This gives

fn(x)

fn(x) + fn2(x)

= fn1

fn1(x) + fn+1(x)

.

We write this as

fn1(x) + fn+1(x)

fn(x)=

fn2(x) + fn(x)

fn1(x).

Using induction, we get

fn1(x) + fn+1(x)

fn(x)=

f0(x) + f2(x)

f1(x).

Observe that

f2(x) =f21 (x) 1

f0(x)= x2 1.

Hencefn1(x) + fn+1(x)

fn(x)=

1 + (x2 1)x

= x.

Thus we obtainfn+1(x) = xfn(x) fn1(x).



3/6

Since f0(x), f1(x) and f2(x) are polynomials with integer coefficients,induction again shows that fn(x) is a polynomial with integer coeffi-cients.

Note: We can get fn(x) explicitly:

fn(x) = xn

n 1

1

xn2 +

n 2

2

xn4

n 3

3

xn6 +

4. LetABC be a triangle. An interior point P of ABC is said to be good ifwe can find exactly 27 rays emanating from P intersecting the sides ofthe triangle ABC such that the triangle is divided by these rays into 27smaller triangles of equal area. Determine the number of good pointsfor a given triangle ABC.

Solution: Let P be a good point. Let l ,m,n be respetively the numberof parts the sides BC, CA, AB are divided by the rays starting fromP. Note that a ray must pass through each of the vertices the triangleABC; otherwise we get some quadrilaterals.

Let h1 be the distance of P from BC. Then h1 is the height for all thetriangles with their bases on BC. Equality of areas implies that allthese bases have equal length. If we denote this by x, we get lx = a.Similarly, taking y and z as the lengths of the bases of triangles on CAand AB respectively, we get my = b and nz = c. Let h2 and h3 be thedistances of P from CA and AB respectively. Then

h1x = h2y = h3z =2

27,

where denotes the area of the triangle ABC. These lead to

h1 = 227

la

, h1 = 227

mb

, h1 = 227

nc

.

But2

a= ha,

2

b= hb,

2

c= hc.

Thus we geth1ha

=l

27,

h2hb

=m

27,

h3hc

=n

27.

However, we also have

h1ha =

[P BC]

,h2hb =

[P CA]

,h3hc =

[P AB]

.

Adding these three relations,

h1ha

+h2hb

+h3hc

= 1.

Thusl

27+

m

27+

n

27=

h1ha

+h2hb

+h3hc

= 1.



4/6

We conclude that l + m + n = 27. Thus every good point P determinesa partition (l ,m,n) of 27 such that there are l, m, n equal segmentsrespectively on BC, CA, AB.

Conversely, take any partition (l ,m,n) of 27. Divide BC, CA, AB re-spectively in to l, m, n equal parts. Define

h1 =2l

27a

, h2 =2m

27b

.

Draw a line parallel to BC at a distance h1 from BC; draw another lineparallel to CA at a distance h2 from CA. Both lines are drawn suchthat they intersect at a point P inside the triangle ABC. Then

[P BC] =1

2ah1 =

l

27, [P CA] =

m

27.

Hence

[P AB] =n

27.

This shows that the distance of P from AB is

h3 =2n

27c.

Therefore each traingle with base on CA has area

27. We conclude that

all the triangles which partitions ABC have equal areas. Hence P is agood point.

Thus the number of good points is equal to the number of positiveintegral solutions of the equation l + m + n = 27. This is equal to

26

2

= 325.

5. LetABC be an acute-angled triangle, and let D, E, F be points on BC,CA, AB respectively such that AD is the median, BE is the internalangle bisector and CF is the altitude. Suppose F DE = C, DEF =A and EF D = B. Prove thatABC is equilateral.

Solution: Since BF C isright-angled at F, we haveF D = BD = CD = a/2. Hence

BF D = B. Since EF D =B, we have AF E = 2B.Since DEF = A, we also getCED = 2B. Applying sinerule in DEF, we have

DF

sin A=

F E

sin C=

DE

sin B.



5/6

Thus we get F E = c/2 and DE = b/2. Sine rule in CED gives

DE

sin C=

CD

sin( 2B) .

Thus (b/ sin C) = (a/2sin B cos B). Solving for cos B, we have

cos B =a sin c

2b sin B

=ac

2b2

.

Similarly, sine rule in AEF gives

EF

sin A=

AE

sin( 2B) .

This gives (since AE = bc/(a + c)), as earlier,

cos B =a

a + c.

Comparing the two values of cos B, we get 2b2 = c(a + c). We also have

c2 + a2 b2 = 2ca cos B = 2a2c

a + c.

Thus

4a2c = (a + c)(2c2 + 2a2 2b2) = (a + c)(2c2 + 2a2 c(a + c)).This reduces to 2a33a2c + c3 = 0. Thus (a c)2(2a + c) = 0. We concludethat a = c. Finally

2b2 = c(a + c) = 2c2.

We thus get b = c and hence a = c = b. This shows that ABC isequilateral.

6. Letf : Z Z be a function satisfying f(0) = 0, f(1) = 0 and(i) f(xy) + f(x)f(y) = f(x) + f(y);

(ii)

f(x y) f(0)f(x)f(y) = 0,for all x, y Z, simultaneously.

(a) Find the set of all possible values of the function f.

(b) If f(10) = 0 and f(2) = 0, find the set of all integers n such thatf(n) = 0.

Solution: Setting y = 0 in the condition (ii), we getf(x) f(0)f(x) = 0,

for all x (since f(0) = 0). Thus eitherf(x) = 0 orf(x) = f(0), for all x Z.Now taking x = y = 0 in (i), we see that f(0) + f(0)2 = 2f(0). This shows



6/6

that f(0) = 0 or f(0) = 1. Since f(0) = 0, we must have f(0) = 1. Weconclude that

either f(x) = 0 or f(x) = 1 for each x Z.This shows that the set of all possible value of f(x) is {0, 1}. Thiscompletes (a).

Let S = n Zf(n) = 0. Hence we must have S = n Zf(n) = 1 by(a). Since f(1) = 0, 1 is not in S. And f(0) = 1 implies that 0 S. Takeany x Z and y S. Using (ii), we get

f(xy) + f(x) = f(x) + 1.

This shows that xy S. If x Z and y Z are such that xy S, then(ii) gives

1 + f(x)f(y) = f(x) + f(y).

Thus

f(x) 1f(y) 1 = 0. It follows that f(x) = 1 or f(y) = 1; i.e.,either x S or y S. We also observe from (ii) that x S and y Simplies that f(x

y) = 1 so that x

y

S. Thus S has the properties:

(A) x Z and y S implies xy S;(B) x, y Z and xy S implies x S or y S;(C) x, y S implies x y S.

Now we know that f(10) = 0 and f(2) = 0. Hence f(10) = 1 and 10 S;and 2 S. Writing 10 = 2 5 and using (B), we conclude that 5 S andf(5) = 1. Hence f(5k) = 1 for all k Z by (A).Suppose f(5k + l) = 1 for some l, 1 l 4. Then 5k + l S. Chooseu

Z such that lu

1 (mod 5). We have (5k + l)u

S by (A). Moreover,

lu = 1 + 5m for some m Z and(5k + l)u = 5ku + lu = 5ku + 5m + 1 = 5(ku + m) + 1.

This shows that5(ku + m) + 1 S. However, we know that 5(ku + m) S.By (C), 1 S which is a contradiction. We conclude that 5k + l S forany l, 1 l 4. Thus

S =

5kk Z.

-00000-


inmo-2012 previous year question papers of indian national mathematical olympiad (inmo) with...

Documents